THE EFFECT OF FOREIGN DIRECT INVESTMENTS O N
FIRM PERFORMANCE IN UKRAINE
by
Oleksandr Talavera
A thesis submitted in partial fulfillment of the requirements for the
degree of
Master of Arts
Kyiv-Mohyla Academy
2001
Approved by ___________________________________________________ Chairperson of Supervisory Committee
__________________________________________________
__________________________________________________
__________________________________________________
Program Authorized to Offer Degree _________________________________________________
Date _________________________________________________________
Kyiv-Mohyla Academy
Abstract
THE EFFECT OF FOREIGN DIRECT INVESTMENTS ON
FIRM PERFORMANCE IN UKRAINE
by Oleksandr Talavera
Chairperson of the Supervisory Committee: Professor Serhiy Korablin, Institute of Economics Forecasting
at Academy of Sciences of Ukraine
All countries are eager to attract as much foreign investments as possible. At
the same time FDI may have not only positive, but also negative effects on
the economy. Positive effects are associated with technology transfer, efficient
allocation of resources, and training of domestic workers. At the same time
entrance of foreign firms could lead to decrease of labor productivity at
domestic firms, which is a negative effect. The main purpose of the paper is
to estimate direct and indirect effects of FDI. First, the research tests for
direct influence of foreign direct investments on firm’s performance, which is
estimated as labor productivity and export. FDI notably increases both labor
productivity and export volumes. Second, we look for spillover or indirect
effects. There is statistical evidence that level of FDI in certain region-
industry increases non –FDI firms performance indicators measured by labor
productivity and volumes of export.
TABLE OF CONTENTS
Table of contents................................................................................................i List of tables ......................................................................................................ii Acknowledgments ............................................................................................iii Glossary ...........................................................................................................iv 1. Introduction...................................................................................................1 2. Theoretical background..................................................................................4
2.1. Literature review....................................................................................4 2.1.1. Direct effects of FDI ....................................................................4 2.1.2. FDI spillovers...............................................................................6
2.2. Model development.............................................................................11 2.3. FDI in transition countries...................................................................15 2.4. FDI in Ukraine....................................................................................18
3. Empirical part ..............................................................................................22 3.1. Data description..................................................................................22 3.2. Econometric models used....................................................................26 3.3. Analysis of results................................................................................31
4. Conclusions and suggestion for future research.............................................36 Works cited .......................................................................................................1 Appendices........................................................................................................3
A1. Graphs..................................................................................................3 A2. Stata 6.0 do-file program........................................................................4 A3. Hausman specification tests. ..................................................................5
A3.1. Hausman specification test for Model 1.........................................5 A3.2. Hausman specification test for Model 2.........................................6 A3.3. Hausman specification test for Model 3.........................................7 A3.4. Hausman specification test for Model 4.........................................8
A4. Questionnaire. Total information about enterprise .................................9
ii
LIST OF TABLES
Number Page Table 1. Statistic characteristics of variables used in research. ...........................23 Table 2. Region distribution of firms...............................................................24 Table 3. Industry distribution of firms.............................................................24 Table 4. Ownership distribution of firms.........................................................25 Table 5. Regression results for FDI influence on labor productivity. ................31 Table 6. Regression results for FDI influence on export share .........................33 Table 7. Regression results for spillovers influence on labor productivity..........34 Table 8. Regression results for spillovers influence on volumes of export.........35
iii
ACKNOWLEDGMENTS
I wish to express gratitude towards Prof. Lutz for his insightful suggestions
and guidance through this thesis writing. I am also grateful to Inessa Love
of Columbia University and Prof. Lehman for their extremely helpful advice
in the empirical part. I also want to express my appreciation to Prof.
Konieczny and Prof. Gardner for comments on the theoretical part and to
EERC Research Center to providing the data. Finally, I wish to thank all
EERC MA students, and in particular Julia Demyanyk, Dmytro Ostanin and
Yuriy Gorodnichenko for their help and support during this thesis
development.
iv
GLOSSARY
Economies of scale: reduction in minimum average costs resulting from through increases in the size (scale) of plant or equipment.
Foreign Direct Investments: all kinds of valuables that are directly invested by foreign investors into objects in order to receive profits (revenues) or reach social effect.
Herfindahl-Hirschman Index: the sum of the squared market shares of each firm in the industry
Home country: a country where the base office of the international corporation is settled.
Horizontal merger: a merger of firms that compete within the same industry combine
Host country: a country, where international corporation establishes its subsidiaries.
Hostile takeover: a change in the ownership of a corporation despite opposition by the original managers or owners.
Joint venture: joint ventures are enterprises established primarily to pool assets of different owners and are of mixed-type ownership.
Investment decision. The decision to build, buy, or lease plant equipment or to start or expand a business.
Multinational corporation (MNC). A company with operations in several countries, headquarters of which is owned by the capital of more than one country and its subsidiaries situated in different countries.
Vertical merger: a firm buys its supplier or vice versa.
S e c t i o n 1
1. INTRODUCTION
Attracting of Foreign Direct Investments (FDI) is one of the most essential
issues in reformation and modernization of the Ukrainian economy. Due to a
substantial technological lag in comparison to developed countries, Ukraine
needs foreign capital that could provide new technologies, new methods of
management and also promote development of domestic investments. The
experience of developed countries shows that investment boom starts with
adaptation of new technologies, brought with foreign capital. Dyker (1999)
emphasizes the ways, through which FDI improve economic performance of
the host countries:
• Integration of host country economy into global economy; • Increase in the aggregate level of investment • Transfer of hard technology (technology of product and process) • Transfer of soft technology (management, marketing methods) • Networking and subcontracting with domestic firms
At the same time, Ukrainian level of FDI per capita is far below that of some
transition countries, in particular the Czech Republic, Hungary or Poland. For
instance, only the USA invested 10 times more in the Polish economy than in
the Ukrainian one1. Such negligible volumes of FDI could be explained by
discouraging investment climate, presently created in Ukraine, and also
suspicious attitude toward foreign investors from government and managers
of some enterprises. Some investors think countries ex ante do not want
attract FDI. “… CEE countries were also unwilling to attract too much FDI.
In transition economies, FDI has typically meant not green field investment,
but the purchase of existing assets, usually during privatization of state-owned
enterprises. Selling state assets to foreigners is often seen as selling the family
1 From the presentation of the US ambassador Steven Pifer in NaUKMA, 2000.
2
silver and encounters widespread political resistance”, (Sinn and
Weichenrieder, 1997).
At the same time, Ukraine has a substantial economic potential, which is
utilized adequately. The main reasons for investing in Ukraine are:
• lack of competition from the domestic firms • cheap labor • potentially large consumer market
Despite these advantages, foreigners are reserved about investing in Ukraine.
Nowadays, the Ukrainian economy really needs inflows of foreign capital,
because of suspension of investment financing from government budget and
the lack of enterprises funds. Among other problems the following should be
emphasized:
• poor legislative framework • unanticipated changes in taxation • equipment deterioration • political instability
All (mentioned above) leads to Ukraine being ranked “B-“ by Moody’s
Company (Infobank, 2001), which is one of the lowest ranks in Europe.
Nevertheless, Ukraine still attracts FDI through the following activities:
• creation of joint venture firms (including sale of stock to foreigners), • creation of free economic zones
While attractiveness of FDI is an important issue for every country we should
not forget about different spillovers that FDI could cause. As a rule, FDI
gives raise to positive externalities. However, we cannot unambiguously assert
these effects of FDI in transition economies, and in Ukraine in particular. As
a rule, transition changes the way economy operates, leading to unexpected
results. Therefore FDI can bring both positive and negative externalities.
Negative spillovers could in the form of a raise in monopoly power of MNCs,
which in order to avoid competition from a Ukrainian firm, acquire and close
it.
In our paper we examine the effects on technology transfer and spillovers
deriving from FDI intensity. More specifically, we survey two problematic
3
questions, using unpublished Ukrainian micro data. Firstly, do establishments
with FDI differ in terms of performance level? Secondly, are there any
beneficial spillovers on firms that have not received FDI? We anticipate that
foreign establishments have comparatively higher levels of performance and
domestic establishments benefit from spillovers.
The data used in this research consist of 292 firm characteristics for 1998-99
and cover Odessa, Kyiv, Kharkiv and Lviv regions. The firms belong to 7
industries according to specification of EERC Research Center2.
We tackle the issue econometrically using panel data technique. The first and
second model tests whether FDI influence on labor productivity and export is
positive or not. According to the obtained results, firms with FDI have higher
labor productivity and volumes of export. In the third and forth models we
tests firms benefits from FDI vicinity. We find a positive spillover effect on
labor productivity and export volumes for non-FDI firms.
The paper is organized as follows. The next section overviews theoretical
background of the work that includes literature review, model development
and issues on the role of FDI in transition countries and Ukraine. In section 3
we describe data, econometric models and discuss the results. Conclusions
and suggestion for future research are in the last section.
2 I do thank EERC Research Center for providing the data.
4
S e c t i o n 2
2. THEORETICAL BACKGROUND
2.1. Literature review
Industrial Organization theory provides us with different approaches to
studying the direct and indirect effects of FDI on host countries. Direct
FDI effects measure the difference in firm performance between firms with
and without FDI. Indirect effects are spread through specific channels and
“examine different aspects of the interaction between MNCs and host
country residents that are plausibly related to FDI spillovers” (Blomström,
Globerman and Kokko, 1999).
2.1.1. Direct effects of FDI
Estimating direct effects, Blomström (1989) investigates differences in
labor productivity, capital-labor ratio, wage level and profitability of firms
with and without FDI. He finds that “… foreign subsidiaries in general
exhibit higher labor productivity and capital intensity than Mexican
manufacturing units of a similar size at the same four-digit industry. Foreign
firms also seem to pay higher wages.” This is explained by a higher labor
and capital quality at foreign companies. However, such indicators as the
share of labor remuneration in value added and profits per unit of capital are
lower in foreign firms. Blomström explains it by the fact that foreign
companies cover their profits to avoid some taxes. Finally, pointing to the
imperfection of his data, he concludes “Although our results indicate
differences in performance between foreign and domestic production units
in the Mexican manufacturing industry, we are unable to show that these
hypotheses are significantly different from zero”.
5
Ponomareva (2000) studies direct FDI effects in Russia focusing on whether
FDI firms perform better than domestic ones. She uses the datasets from
State Statistic Committee of Russia mostly on energy, fuel and foodstuff
production industries. A number of firms are situated in the central region,
near natural resources or metal processing centers. She develops an
econometric model where output depends on employment, capital,
economy of scale, minimum efficiency scale, existence of FDI, industry and
region. The author’s does not confirm a significant influence of foreign
majority ownership, so that firms with the prevailing share of foreign capital
do not show a better performance. Furthermore, the study finds that output
in plants with FDI is higher than that of domestic ones.
Similar research is conducted by Konings (2000) for Poland, Bulgaria and
Romania. He finds that “foreign firms perform better than firms without
foreign participation only in Poland. In Bulgaria and Romania, no robust
evidence is found of positive foreign ownership effect”. The author explains
this by the time lag needed by firms to restructure and effect on
performance productivity. According to Konings, Poland is advanced in the
transition process comparing to Bulgaria and Romania.
In another study at the macro-level Mykytiv (2000) analyzes influence of
FDI on economic growth in Ukraine is estimated. The author agrees that
FDI is commonly linked to the technological progress in the country,
because of the transfer of new technologies and inputs in innovations. This,
in turn, is the basis of argument that enterprises with foreign investments
exhibit a higher labor productivity comparing to domestically owned
manufacturing. Mykytiv also argues that FDI is positively correlated with
exports, since foreign investors often adopt export-oriented policy. It
prompts competition among local enterprises and spreads new competitive
technologies. Mykytiv (2000) develops a simultaneous model for the
economy of Ukraine and uses the Error-Correction Model to explain long-
term trends. However, he finds no significant results for the FDI influence
on economic growth. These neutral results may be the result of inadequate
country statistics.
6
2.1.2. FDI spillovers.
Technology Transfer
Describing indirect FDI effects, Blomström and Kokko (1997) discuss
transfer and technology diffusion from multinational companies to host
countries, as well as prevalent ownership of commercial technologies by
multinational companies. They consider theoretically the main technology
transfer channels such as:
• Contribution to efficiency of domestic firms • Introduction of new know-how • Transfer of techniques for inventory and quality control
Glass and Saggi (1998) develop a model in which international technology
transfer occurs through different channels. They evaluate the role that FDI
plays in technology transfer promotion. FDI is the most important channel
of international technology transfer. They argue that a faster flow of FDI to
the host country increases the rates of innovation, imitation and
international technology transfer. They also emphasize imitation as a source
of technology transfer. Finally, they suggest that rates of innovation and
imitation remain the same and FDI generate level effect only.
The technology transfer channel is also theoretically analyzed by Blomström
(1987). He concludes that “… such a transfer is a central activity of MNCs,
and this may stimulate domestic firms to hasten their access to a specific
technology”.
Ponomareva (2000) examines the impact of technological spillovers from
FDI on domestic enterprises. She mentions a positive effect from FDI
spillovers and concludes, “The effects [of FDI spillovers] depend on host
country and host industry characteristics and the policy environment in
which the multinationals operate”. The author mentions an intra-region
transfer of know-how and technology and finds that “domestic firms
located nearby multinationals benefit from this vicinity”.
7
Kinoshita (2000) examines the effect of technology diffusion from FDI in
explaining the total factor productivity growth. She uses unpublished firm-
level data of the Czech manufacturing sector for period of 1995-98. She
finds that both foreign joint ventures and foreign presence in the sector do
not have significant effects on productivity. The author finds that “the rate
of technology spillovers varies greatly among sectors. In oligopolistic sectors
such as machinery, there exists a significant rate of spillovers from having a
large foreign presence” and no spillovers in more competitive food, textile,
wood and chemical industries.
Competition
As for competition effects, Blomström (1987) describes it as an increase in
competition when multinational companies enter the host country markets.
According to Blomström, greater competition leads to a more efficient
market structure. The author argues that the Herfindahl index could be a
proxy for the estimation of this effect. However, this explanatory variable is
not significant. He explains this by the fact that “due to underdevelopment,
Mexico is a relatively small economy, but a highly protected one”. In other
words, domestic firms are legally protected from loosing their market shares
by MNCs.
Blomström and Kokko (1997) also examine the influence of international
companies on the performance of the host country, as well as the effects on
competition and industry structure in the host countries. They conclude that
FDI contributes to productivity growth and exports in host countries, yet
the exact nature of correlation between foreign and domestic firms could
vary among industries and countries: some industries are more protected by
government, some are less protected.
Ponomareva (2000) stresses the fact that competition with foreign firms
forces domestic companies to protect their market share and profits. In
contrast to the previous study, she finds negative effects. She concludes,
“increases in foreign ownership negatively affect the productivity of wholly
domestically owned firms in the same industry”.
8
Similarly, Konings (2000) finds no spillovers effects in Poland, but there are
negative spillovers in Bulgaria and Romania. He explains that “… increased
competition from FDI dominates technological spillovers to domestic firms.
It suggests that inefficient firms will loose market share due to foreign
competition, which in long run should increase the overall efficiency of an
economy”.
Furthermore, Kinoshita (1998) tries to decompose spillover effect into
competition, training, foreign linkages and demonstration effects3. The
author uses the survey data for 468 manufacturing firms in China between
1990-1992. She finds that the catch-up effect (spillovers) is “more important
for domestically owned firms than for foreign firms, which rely on the
import of intermediate goods”. The author concludes that “Chinese local
firms have survived increased competition due to the entry of more
advanced foreign firms and have accomplished rapid growth because of
this”.
Training of labor and management
Blomström (1989) mentions that the training spillover channel can be a
result of worker training by foreigners investing in human capital that
spreads not only on foreign but also on domestic firms. “In Mexico ... many
managerial people in large locally owned firms started their career in a
MNC, and management practices may in this way be substantially improved
in domestic firms”. However, he does not estimate this hypothesis
empirically.
Moreover, Kinoshita (1998) finds worker training an important source of
productivity growth. However, it has some particularities. Domestic firms
being afraid to loose their market shares, train their workers. Kinoshita
suggests, “This might have facilitated the process of intra-industry spillovers
from foreign investments”. At the same time foreign-owned firms are
unlikely to invest in the education of local workers. On the contrary, they
3 See findings on training, foreign linkages and demonstration effects below
9
“tend to maintain product quality by improving intermediate goods from
their home countries and by transferring managers from their
headquarters”.
Foreign linkage effect
Blomström and Kokko (1997) distinguish backward linkage and forward
linkage effects. A backward linkage occurs during interaction between
multinational companies’ branches with suppliers. The authors suggest that
backward linkage is associated with MNCs assistance in establishing
production facilities by suppliers, increasing quality of raw materials and
training of management. A forward linkage is associated with consumer-
MNC relationships. This channel is less evident than the previous one, and
Blomström and Kokko mention insignificance of the forward linkage effect.
Similarly, Kinoshita (1998) also included foreign linkages proxy, but finds
statistically insignificant coefficients.
Demonstration effect
Transferring of new technologies and innovations could be adopted via
simulating them. According to Blomström, Globerman and Kokko (1999),
“The successful introduction of new production techniques and new
products reduces the subjective risk surrounding the adoption of the
innovation and should, therefore promote its adoption more widely
throughout the population of potential adopters in the host country”
Blomström and Kokko (1997) suggest it as an important channel of
spillovers. They suggest, “…pure demonstration effects often take place
unconsciously … and often intimately relates to competition”.
Kinoshita (1998) determines the demonstration-imitation effect: when
domestic firms observe activity of their multinational firms they start to
imitate or copy in order to become more productive.
10
Thus, we can see that the issue of FDI and spillovers is highly appealing for
research. FDI has direct and indirect impacts. Direct FDI effects measure
difference in firm indicators between firms with and without FDI. Indirect
effects are spread through specific contacts between MNCs and domestic
firms. There are five main indirect effects found in relevant literature. The
technology transfer effect appears when domestic firms receive new
technologies and know-how for lower costs from MNCs. The catch-up
effect simply means that foreign firm catches up the share of local market or
domestic firm looses its market share. The competition effect arises when
entrance of foreign firms forces domestic firms to act more efficient in
order to save their profits and shares. The foreign linkage effect appears
when foreign owned companies use services supplied by local firms. The
training effect is a situation when foreign firms provide training for their
workers and managerial staff, which in future can be hired by domestic
firms.
11
2.2. Model development
As we can see from the previous section, there are two main subtopics of
the FDI issue:
• FDI influence on firm’s performance • Spillover estimation
Recently developed empirical models estimate two effects either
simultaneously [Ponomareva (2000), Konings (2000)] or separately
[Kinoshita (1998), Blomström (1989)].
One of the earliest econometric models for estimation of FDI influence was
developed by Magnus Blomström (1989) in his Mexican manufacturing
sector research. He estimates labor productivity as an indicator of firm’s
performance4. The model is:
)),,(,,,,( 21 FSLQLQADSCALEHKLfy ddd = , where
dy - value added in domestically owned private plants divided by total employees in these plants. dKL - the ratio of total assets to the total number of employees
dSCALE - measure of scale AD - average effective working day
1LQ - ratio of white-collar workers to blue-collar workers
2LQ - measure of labor quality H – Herfindahl index FS – share of employees in industry’s total employment in foreign plants
Among factors influencing firm performance Blomström uses capital
intensity, quality of labor force, market structure, economy of scale and
foreign presence. The last is estimated as the share of employees in an
industry employed in foreign plants. As a proxy of capital intensity
Blomström employs the ratio of total assets (book value) to the total
number of employees in the domestically owned plants. Moreover, he
suggests, “Labor productivity may also differ across plants because of scale
4 Blomström would prefer to use the ratio of net output to an index of total factor inputs, but cannot do
it because of unavailable data.
12
economies”. Diseconomies of scale are represented by the ratio of the gross
production of the largest plants in industry to gross production of an
average privately owned Mexican plant. The quality of labor force is
estimated by two variables. First, it is a ratio of white-collar to blue-collar
workers for an industry. Second, it is the error term “e” in the regression
eFSbaLQ ++= *1. The Herfindahl index, which is used as a measure
of concentration in market structure, is calculated as the sum of the squared
market share of all firms in the industry. Variable AD corrects for the
possibility of systematic differences in holidays, strikes, etc., in order to
receive a better estimation of labor productivity.
A similar approach is developed by Ponomareva (2000) in her research on
FDI spillovers in Russia. She estimates both FDI and spillover effect
simultaneously. She also uses firm’s output as a performance indicator,
which depends on capital, labor, economy of scale, FDI spillovers, FDI
presence and minimum efficiency scale. The model is:
ijijij
jijijij
espillafdiascalea
scmefacapaempaaout
++++
++++=
654
3210
*
_*)ln(*)ln(*)ln(
where, i – firm index
j – industry index out – output emp – employment cap – capital scale – economies of scale mef_sc – minimum efficiency scale fdi – dummy variable for FDI spill – spillover variable e – error term
Capital stands for fixed assets at the beginning of the year. Labor denotes
average annual employment. Like in the previous model, Ponomareva uses a
scale variable, but it is measured as an establishment’s production over the
average production in the industry. Minimum efficiency scale is estimated as
median output over average output in the industry. This variable
characterizes the distribution of firms in the industry. The FDI dummy
13
variable is valued 1 if a firm has FDI and 0 otherwise. Finally, a spillover
variable is estimated as the share of output in an industry accounted for the
foreign firms in total output.
Like Ponomareva in the previous model, Konings (2000) also estimates
both FDI and spillovers effect. He constructs a similar log-linear production
function:
ittittitititiit SpillFDIFDImalky εαηααηαααα ++++++++= 7654311
where, i – firm index, t – year index.
yit – log output kit – log of capital lit – log employment mit – log of material inputs
tη - time varying factor FDIi – share of firm hold by foreigners FDI tη – interaction of foreign ownership and time trend spill – sector level spillover variable e – error term
Compared with the previous model, Konings adds log of material inputs.
He also includes a time varying factor, which measures common aggregate
shocks in production for example technological progress or other
unobserved factors, and an interaction variable of foreign ownership and a
time trend variable, which proxies foreign ownership influence on both level
and growth of productivity. For the FDI variable Konings uses the fraction
of shares held by foreign investor, and spillover variable is represented by
the share of output accounted for by foreign firms in total output at the
two-digit NACE sector level.
There is also research that estimates spillover effects separately. Magnus
Blomström(1989) supports the idea that “[t]echnical progress can be studied
by observing changes in the best-practice technology between two periods.
The more rapid the technical progress has been, the faster the frontier has
moved”. That is why he uses relative changes in labor productivity in the
14
best practice plants within each industry between 1970 and 1975 as a
representative for technological growth. The model is:
),,,( FSMGHyfe ∆= ,
where y∆ - technical progress
H – Herfidahl index MG – market growth FS – foreign ownership share e – efficiency index
Blomström defines the market growth variable as the relative growth of
employment in each industry. He identifies a company under foreign
ownership if at least 15 percent of a company is foreign owned. Finally, the
Herfidahl index represents market structure. The dependent variable e, is
estimated in two stages. “First, the efficiency frontier is obtained by
choosing the size class within each four-digit industry showing the highest
value-added per employer”. Second, Blomström finds the ratio of industry
average5 to the value found at the first stage.
Thus, previously developed models, which estimate influence of FDI and its
spillovers on firm’s performance, can be subdivided into all-effects
estimations and separate-effects estimations. In most models, change in
added value or in output depend on FDI or spillover variable and on
additional control variables such as capital, labor, economy of scale, quality
of labor force, minimum efficiency scale, etc. Having considered mentioned
above, we develop our own models, which are described in section 3.
5 Estimated as ratio of total value added in each industry to the total number of employees.
15
2.3. FDI in transition countries
Transition countries are characterized by a need for foreign investment,
especially FDI. Below we describe the main features of investment climates
in Poland, Czech Republic, Hungary and Russia.
The Czech Republic differs from some Eastern European countries by
macroeconomic stability, well qualified labor and anticipated political
environment. According to CzechInvest6, the Czech agency for attraction
FDI, all sectors of economy are opened to a foreign presence. The Czech
government developed a standard package of incentives for manufacturing
investments in 1998. Incentives are offered for investors who invest $10
million in manufacturing through the creation of a new firm. The package
consists of duty free import of equipment, delays in value added tax
payment, training grants and additional incentives after reinvesting.
Moreover, the Czech government has created a sophisticated legal
environment based on commercial code, banking law and tax code.
According to CzechInvest data, the Czech Republic attracted the total of
US$19.3 billion of FDI in 1991-1999. Germany was the largest foreign
investor and contributed US$5.0 billion (26.2 %). The second largest
investor was Netherlands with US$4.6 billion (24.0 %). Austria and the USA
follow with US$2.3 billion (11.8%) and US$1.7 billion (9.0%), respectively.
The introduction of investment incentives resulted in a significant increase
of FDI in 1998. As for industrial allocation, the most attractive are financial
services (US$3.0 billion or 15.8%), wholesale trade (US$2.7 billion or
13.8%), non-metallic mineral products (US$1.5 billion or 8.0%) and post
and telecommunications (US$1.2 billion or 6.4 percent).
In Poland foreign investments also play a considerable role. According to
the U.S. Department of State7, all political parties and social groups support
any actions for FDI attractions in all spheres of the Polish economy except
6 Available at web-site http://www.czechinvest.org
7 Available at web-site http://www.state.gov
16
agricultural land. The Polish government has approved a comprehensive
legal framework that protects property rights and investments, provides
equal treatment for both domestic and foreign firms and allows repatriation
of profits abroad. Legislative environment is based on Commercial Code,
Law on Economic Activity, The Law on Companies with Foreign
Participation and others. The amount of FDI attracted in Poland has being
increasing since 1991 (PAIZ, 2000). For instance, cumulative FDI in 1998
was $27,279.6 million, and in 1999 - $35,170.8 million. The reasons are the
capacity of Polish markets, skilled work force and low labor costs. Total
FDI in Poland reached USD 38.9 billion in 2000. German and US
companies invested US$ 6,007.3 million (17.3 %) and US$ 5,152.9 million
(14.7%), respectively. Other large investors are France (11.1%), the
Netherlands (9.2%) and Italy (9.1%). The most attractive industries for
investing are financial services (22.4%), food processing (13.1%),
transportation equipment (12.5%), trade and repairs (9.7%), non-metal
goods (5.9%), construction (5.5%) and transport, storage and
communications (5.4%)
According to the U.S. Department of State, Hungary attracted over US$ 23
billion of FDI from 1989 to 2000, which is a significant share of all FDI
invested in Central and Eastern Europe during this period. The current
economic and legal environment encourages FDI in all areas of the private
economy. There are four main types of FDI in Hungary: establishment of a
new business, joint venture; privatization and portfolio investment, or
participation in capital raising. The major investing countries are USA,
Germany, France, Austria, and the Netherlands, followed by Italy, Sweden,
Great Britain, Switzerland, Japan, and Canada. In 1999, 55 percent of all
FDI was invested in manufacturing, followed by telecommunications (15%),
energy (13%), banking/finance (6%), and commerce (6%).
According to the U.S. Department of State, “the Russian economy has
shown strength recovering from the August 1998 financial crisis, and real
growth in the economy has helped spur limited new investment from both
domestic and foreign investors. Many problems persist, however, including
17
chronic difficulties in the overall investment climate and a weak commercial
banking sector. President Putin’s government has shown a strong interest in
attracting foreign investment and has promised to enact structural changes
that would improve the environment for investors. However, most of these
key steps have yet to be enacted.”
Among the different problems existing in Russia, crime is one of the most
viable concerns of foreign and domestic businesses, particularly those
dealing with large amounts of cash and goods. “Much crime is tied to
commercial activity, and many Russian entrepreneurs report that they must
pay kickbacks and protection to stay in business. Furthermore, foreign
investors have identified corruption as a pervasive problem, both in the
number of instances and in the size of bribes sought.” (U.S. Department of
State).
Cumulative FDI from 1992 to 1999 totals about $US 16 billion. Among the
largest investors are the following countries8: Germany - $US 6,946 million
(23.7%), USA - $US 6,349 million (21.7%), UK - $US 3,628 million (12.4%),
Cyprus - 3,440 million (11.8%) and France - $US 3,249 (11.1%).
The most attractive for FDI Russian industries in 1999 were: the fuel
industry $US 1,700 million (17.8%), trade and catering $US 1,622 million
(17.0%), consulting services $US 1,481 million (15.5), food industry $US
1,415 million (14.8), transport and communication $US 521 million (5.5%).
Eastern and Central European countries could roughly be subdivided
according their FDI attractiveness as more attractive (Poland, Czech
Republic and Hungary) and less attractive (Russia, Ukraine). Thus,
geographic factor also plays a great role in investor’s decision.
8 In cumulative terms
18
2.4. FDI in Ukraine
The collapse of the Soviet Union in 1991 created 15 independent states.
One of those countries is Ukraine, which had a land area of 604,000 square
kilometers (232,000 square miles) and a population of about 50 million. “In
the 1980s, Ukraine produced 16-18 percent of Soviet industrial output and
23-24 percent of Soviet agricultural output: in 1989 it produced 34 percent
of Soviet steel, 23.5 percent of coal, 46 percent of iron ore, 56 percent of
sugar and 36 percent of TV”, (Yegorov, 1999).
It was supposed that Ukraine be highly attractive for FDI. Ukraine had
cheap but at the same time skilled labor and available row materials.
However, investors were not in a hurry. FDI per capita in 1995 in Ukraine
was only US$13, while in Hungary it was US$1017 and in the Czech
Republic US$575.
Despite business recession, equipment deterioration, instability of
macroeconomic situation and other causes foreign investors slowly but
gradually invest in Ukraine, except in areas prohibited by law. Government,
financial intermediaries and firms with FDI actively operate in the
investment market. The government provides formal rules or legislation
environment at both national and local levels, though officials are often
interested in political power and private benefit. As for informal rules,
“...corruption follows directly from the degree of discretion officials have in
granting approvals for private business. Unofficial payments have to be
made at all stages of the licensing and permissions process” (Kudina, 1998).
Another actors in the market are financial intermediaries or security brokers,
who act as agents for investors. They compete at the investment market
supplying consulting services. KINTO Investments and Securities, Alpha
Capital, Dragon Capital are leaders at the market. “The leading Ukrainian
securities company, KINTO Investments and Securities, was formed with a
substantial contribution from Wasserstein Parella of the US and European
Privatization and Investment Corp. In turn, KINTO has created a number
19
of daughter companies, which operate on the Ukrainian equity market”
(Yegorov, 1999).
Foreigners mostly invest through creation of own firms (for example,
Cargill’s new seed processing plant in the Donetsk Region) or buying equity
(For example Irish CRH, which bought controlling interest of cement plant
in the Khmelnitsky Region). Foreign investment has the following forms:
• Foreign exchange • Domestic currency (reinvesting) • Any movable and immovable property • Equity • Corporate rights • Immaterial assets (know-how, software)
Investments in Ukraine are formally regulated by laws and other legal acts.
The following laws should be emphasized:
• The Law “On Foreign Investing Regime”, dated March 19, 1996 • The Law “On Foreign Investments” dated March 16, 1992 • The Law “On State Program of Encouraging Foreign Investments
in Ukraine”, dated December 17, 1993 • Cabinet Resolution on a Foreign Investment Regime, dated May
20, 1993;
The main features of Law “On Foreign Investing Regime” are:
• Registration requirement of foreign investments with local authorities
• Regulation of types of foreign investments • Privileges for foreign investors • Guarantees for profit repatriation • Exemption of custom duties for foreign contributions to statutory
fund.
In order to attract more investors, there are provisions on legislature
changes in Law “On Foreign Investments”:
• Foreign investors are guaranteed protection against changes in legislature for 10 years
• Guarantees against illegal nationalization • Compensation and reimbursement of foreign investors losses
(nationalization)
20
• Guarantees if investment activity is terminated (repatriation of profits and invested assets).
• Guarantees for profit repatriation • State registration and control for foreign investments
The law “On State Program of Encouraging Foreign Investments in
Ukraine” describes ways to attract more FDI into agriculture, light, fuel,
medicine, chemical and transport industries.
However, these laws cannot fully clarify ambiguity in the FDI legal
environment. The government can issue and an unanticipated amendment
in the middle of year, when plans for most companies’ development are
already established. Moreover, numerous cases of corruption and bribery are
apparently not conductive to the attraction of FDI.
According to the official statistical office, Derzhkomstat, FDI in Ukraine
has reached the total of $3.25 billion since 1992. The United States has
invested $589 million and has become the largest investor; the Netherlands
follows with $301 million. Other big investors are Russia ($288 million),
Great Britain ($243 million), and Germany ($229 million)
The most important for FDI industries are machinery (13% of total FDI),
and food industry (21%). Domestic trade also plays a significant role in FDI
attraction (16%) (Derzhkomstat, 2000).
According to the Institute of Reforms (2000) there are many differences in
FDI attractiveness by regions9. Kyiv City is the most attractive region in
Ukraine. It has a comparatively developed transport and communication
systems, infrastructure, financial institutions. The average salary in Kyiv is
twice than the average in Ukraine. As on June 2000, $1.2 billion were
invested in the city, 32,9% from the USA, 16.5% from Cyprus, 9.8% from
Austria, 6.8% - Hungary and 4.2% from Switzerland.
9 The ranking of investment attractiveness is based not only on investment but also on economic,
infrastructure, social regional features.
21
The second in the ranking of investment attractiveness (Institute of
Reforms, 2000) is the Donetsk region. As of July 2000, due to the creation
of free trade zones “Donetsk” and “Azov” and attractive economic
situation investors, contributed to the Donetsk region $293.5 million.
Dnipropetrovsk, Lviv, Zaporizhia and Kharkiv could also be named as
leaders. These regions attracted $184.33 million, $125.85 million, $218.00
million and $88.86 millions, respectively. Next group is the “followers”. It
consists of the Odessa ($190.12 million), Kyiv excluding the city ($302.9
million), Poltava ($211.12 million) regions, and the Crimea Republic ($150.1
million). Each of these regions attracted a significant amount of FDI but
because of various reasons10 lags behind the leading regions. Chernivtsi
($8.16 million), Zakarpattja ($81.73 million), Mykolaiv ($44.87 million),
Lugansk ($30.25 million), Ivano-Frankivsk ($37.64 million), and Kherson
($32.19 million) regions are included in the “main” group. The regions of
this group have variega ted indicators. For example the Ivano-Frankivsk
region has ranked 13th in the stocks parameter, but there is no regional
company listed in PFTS, the Ukrainian index. The group of “outsiders” is
characterized by an undeveloped business infrastructure and consists of
Chernigiv ($46.78 million), Vinnytsja ($17.06 million), Rivne ($44.79
million), Ternopil ($19.65 million), Khmelnitsky ($14.67 million), Sumska
($34.56 million), Kirovograd ($17.96 million), and Cherkasy ($103.60
million).
Thus, on the one hand Ukraine could potentially be attractive to foreign
investors. It is possible to earn huge profits. On the other hand, it is
extremely risky to put money into Ukrainian firms due to political instability
and vulnerable legislature.
10 For example the Odessa region does not have a developed financial infrastructure, its level of
production significantly depends on companies such as Odessa refinery or Ukrtatnafta in the Poltava region.
22
S e c t i o n 3
3. EMPIRICAL PART
As we can see from the previous part, FDI and spillover issue has
substantial theoretical background. Unfortunately, nobody has ever
researched the topic in Ukraine. Having stated the main question of paper
“Does Ukraine benefit from FDI?” we test for direct and indirect FDI
effects on labor productivity and export volumes. In our research we use
unpublished firm level data from the EERC Research Center.
3.1. Data description
The data used in this research consist of two EERC Research Center
datasets. The first includes micro-level information on fixed assets, labor
force, sales, export, import, barter operations, and industry-region
information. The second contains information on FDI presence in certain
firms.
To present such variables as capital we could use different estimations of
fixed assets. According to a theoretical study (Ponomareva, 2000), the
balance sheet value could be the best proxy for capital, since it reflects real
capital capacities of the firm. All data are constant 1998 prices, cobverted
using producer price index from the UEPLAC(2000) web site11 (See Table
1).
There are 292 observations for manufacturing firms for 1998 and 1999.
25% of them have some FDI. A firm is assumed to have FDI when12:
11 Available at http://www.ueplac.kiev.ua
12 Firms unwillingly report on FDI availability. Therefore, the EERC Research Center used questionaires about changes in FDI in order to find the FDI existence in companies
23
• They had changes in FDI during last period • They had foreign ownership13
The dataset covers Lviv, Kyiv, Odesa and Kharkiv regions, which represent
West, Center, South and East of Ukraine, respectively. Regional distribution
with frequencies and percentages is described in Table 2. As can be seen
from the Table 2, the share of Kyiv, Lviv and Kharkiv regions is 30% each,
while the share of Odesa region is 10%. This may be explained by the fact
that Ukrainian South is less industrialized than central or eastern areas, and
the fact of unwillingness of Odesa region’s managers to take part in EERC
survey.
Table 1. Statistic characteristics of variables used in research.
All firms FDI firms Variable
Mean Std. Dev. Mean Std. Dev.
Balance value of fixed
assets, UAH 1998
17324.32 54366.9 5904.55 12853.74
Sales, UAH 1998 5026.05 15245.07 3353.26 7379.38
Import, UAH 1998 902.15 3525.32 1548.95 3914.26
Production, UAH 1998 5169.32 15474.25 3948.94 10837.29
Labor, workers 457 1019 255 508
Export, UAH 1998 852.12 3801.31 1136.95 4246.77
13 We assumed that firms with foreign ownership should have FDI in either material or at least non-
material form.
24
The dataset covers 7 industries. Most of the firms are involved in food
industry (25%) or in metal processing (20%). A large number of firms do
not identify themselves with any particular industry (22%). The industry
distribution of firms is summarized in Table 3.
Table 2. Region distribution of firms.
All firms FDI firms Region
Frequency Percentage Frequency Percentage
Kyiv region 88 30.14 22 30.14
Lviv region 90 30.92 26 35.62
Kharkiv region 89 30.48 22 30.14
Odesa region 25 8.56 3 4.10
Table 3. Industry distribution of firms
All firms FDI firms Industry
Frequency Percentage Frequency Percentage
Metallurgy 24 8.22 5 6.85
Metal processing 58 19.86 8 10.96
Wood and Paper 15 5.14 5 6.85
Construction
materials
26 8.90 5 6.85
Light 30 10.27 9 12.33
25
Food 74 25.34 18 24.66
Others 65 22.26 23 31.51
The ownership structure of available data is depicted in Table 4. A
significant share of firms (36%) constitutes an unspecified form of
ownership14. Workers own 17% of firms in the sample. Other physical
entities are either retired persons or those who bought shares during
certificate auctions.
Some firms with foreign ownership were added to the original dataset in
order to increase the sample of firms with FDI15.
Table 4. Ownership distribution of firms16
Ownership Frequency Percentage
Workers 49 16.78
Managers 13 4.45
Government 7 2.40
Other physical entities 27 9.25
Other Ukrainian companies 29 9.93
Other foreign companies 61 20.89
Other 106 36.30
14 See in appendices: A4. Questionnaire. Total information about enterprise in appendices, question #3.
15 Total amount of added firms is about 45.
16 On the basis on major ownership.
26
3.2. Econometric models used
The main aim of the thesis is to estimate the influence of FDI on firm’s
performance and find the region-industry spillover effect. In order to estimate
the former effect, we develop the following analytical model:
),,,,,,( itiiiiititit ScaleOWNERSHIPFDIREGIONIndustryLKfP = (1)
where i – index for firm and t – index for year
Pit – firm performance, estimated as labor productivity or export volume
Lit – labor that that is the quantity of workers in the firm,
Kit – capital stock or the balance value of fixed assets,
Scale it – economy of scale proxy, estimated as the ratio of firms production to
the average production in industry
INDUSTRYi – industry, one of seven industries according to specification of
the EERC Research Center,
OWNERSHIP i – type of ownership, one of types of ownership according to
specification of the EERC Research Center,
REGIONi – region, where the firm is situated,
FDIi – a dummy variable that shows the existence of FDI.
The dependent variable, i.e. performance, could be estimated in various
ways. The ideal representation is value added or value added per worker.
However, due to data restrictions sales, production, barter, export and
import, we decided to use the Hausman specification test to check for a
correct econometric specification17.
Econometric models used are shown below.
Model 1. Labor productivity is assumed to be a performance indicator and our model is:
17 See results appendices: A3. Hausman specification tests.
27
itii
iiit
it
it
it
OWNOINDUSTRYS
REGIONRFDILK
constLY
ε
αα
δδδ
σσσ
ρρρ
++
++++=
∑∑
∑
==
=
6
1
6
1
3
121 lnln
(2)
where
FDIi, is a dummy variable taking the value 1 if the firm has ever had foreign
direct investments, and 0 otherwise.
REGIONi, INDUSTRYi are dummies, which specify an industry and region,
respectively. For region dummies Odesa region is the base and R1 denotes
Kyiv region, R2 – Lviv region and R3 - Kharkiv region. Unspecified industry is
the base for industry dummies and other dummies are: S1 – metallurgy, S2 –
metal processing, S3 – wood and paper industry, S4 – construction materials
industry, S5 – light industry and S6 – food industry.
OWNoi – are dummies that determine type of ownership. Unspecified kind
of ownership is the base for ownership dummies. We denote O1 – workers
ownership majority, O2 – management, O3 – state, O4 – other physical
entities, O5 – other Ukrainian companies and O6 – other foreign companies.
Our Hypotheses for the model 1 are the following:
H0: á2=0: FDI does not affect labor productivity
H1: á2>0: FDI has a positive influence on labor productivity
We anticipate the rejection of our H0.
Model 2. Performance is measured by export volume. Theoretically, if a firm
exports more, it has comparative advantage, which is a positive fact. The
second model is the same as model 1, but we add economy of scale proxy,
estimated as the ratio of firm’s production to the average production in
industry: We also use capital and labor variable separately instead of labor
productivity.
28
itiiti
iiititit
OWNOScaleINDUSTRYS
REGIONRFDILKconstExp
ε
ααα
δδδ
σσσ
ρρρ
+++
+++++=
∑∑
∑
==
=
6
1
6
1
3
1321 lnlnln
(3)
Hypotheses for model 2:
H0: á3=0: FDI does not affect export volumes,
H1: á3>0: FDI has a positive influence on export
We anticipate the rejection of H0.
Endogeneity is a problem associated with models 1-2. It is typical in Ukraine
that firms with FDI have higher labor productivity and firms with larger
labor productivity attract more FDI. In other words, foreigners make
Ukrainian firms to operate more efficient and the best firms also attract
FDI. The same links can be traced between FDI and export. Firms with
FDI have larger volumes of export and conversely large volumes of export
magnetize FDI.
In order to solve this problem we used the 2 stage methodology. While FDI
is highly correlated with export, the latter, in turn, is not closely correlated18
with labor productivity. Therefore, we construct the following measure:
ititi EXPconstFDIprobit εα ++= ln)( (4)
and then, using the GLS in order to avoid heteroscedasticity problem:
itii
iiit
it
it
it
OWNOINDUSTRYS
REGIONRFDILK
constLY
ε
αα
δδδ
σσσ
ρρρ
++
++++=
∑∑
∑
==
=
6
1
6
1
3
121 lnln
(5)
18 R2 =0.15
29
Thus, we estimated the real effect of FDI on labor productivity. Similarly,
export is estimated as an indicator of firm performance:
itit
iti L
YconstFDIprobit εα ++= ln)( (6)
itiiti
iiititit
OWNOScaleINDUSTRYS
REGIONRFDILKconstExp
ε
ααα
δδδ
σσσ
ρρρ
+++
+++++=
∑∑
∑
==
=
6
1
6
1
3
1321 lnlnln
(7)
We anticipate that FDI has a positive effect on firm’s performance
estimated as labor productivity or export.
In models 3-4 we investigate whether a firm that does not receive FDI
benefits from FDI in other firms in its industry-region. In other words, we
want to estimate the influence of FDI intensity, which is represented as a
share of investment in a certain region-industry, on performance of firms
that do not have FDI.
There is a less potential for “endogeneity”19, as we do not expect the
productivity of firms that do not receive any FDI to be affected by the
proportion of FDI in other firms in their industry-region. It is not likely that
FDI in the industry-region should somehow be correlated with the labor
productivity of firms that do not get any FDI. To control for unobserved
heteroscedasticity we use the GLS in models 3-4.
Model 3. We use the labor productivity as a measure of firm performance,
and the model is:
itiiiit
it
it
it INDUSTRYSOWNOSPILLK
constLY
ελασ
σσδ
δδσδ +++++= ∑∑==
6
1
6
11 lnln (8)
19 I thank Inessa Love from Columbia University for clarifying this point.
30
We do not include regional dummies because of insignificance their
coefficient. The spillover variable is the percentage of FDI in region
multiplied by percentage of FDI in industry of non-FDI firm.
Thus, the Hypotheses for the model 3 are:
Ho: ë>0: FDI intensity causes negative or no externalities
H1: ë�0: FDI intensity increases labor productivity of Non-FDI firms
We anticipate the rejection of H0.
Model 4. We use export as a proxy for firm’s performance. The model is:
itiiiititit INDUSTRYSOWNOSPILLKconstExp ελαασ
σσδ
δδσδ ++++++= ∑∑==
6
1
6
121 lnlnln (9)
Our hypotheses are:
Ho: ë�0: FDI intensity causes negative or no externalities
H1: ë>0: FDI intensity positively influences export volumes of non-FDI
firms
We anticipate the rejection of H0.
31
3.3. Analysis of results
In order to test all four hypotheses we tested 4 models. Our findings for the
hypotheses testing are described in tables 5-8. Model 1 is estimated as
variations of Equation 5. We test for FDI impact on labor productivity.
Table 5. Regression results for the FDI effect on labor productivity.
it
it
LY
ln it
it
LY
ln it
it
LY
ln it
it
LY
ln
constant 3.110221 *** (.3798949)
3.36888*** (.6094581)
3.797782*** (.5813996)
3.361068 *** (.6017184)
it
it
L
Kln -.0321387
(.0954852) -.0958973 (.0990565)
-.0777378 (.0874039)
-.0483727 (.0878096)
FDI .7544024*** (.1440682)
.7314491*** (.1436722)
.8042352*** (.137226)
.7737273*** (.1398052)
Kyiv region .1963453 (.4919964)
.0382721 (.4099749)
.0755219 (.4087269)
Lviv region -.5243072 (.5092676)
-.3578108 (.4218442)
-.3236778 (.4304532)
Kharkiv region
-.00652 (.5054732)
.0584154 (.4184861)
.1697789 (.420667)
Metallurgy industry
-.0532733 (.3458821)
.1002837 (.3539814)
Metal processing
-1.2091*** (.2727451)
-1.105147*** (.2791934)
Wood and paper
.2423589 (.5650992)
.069621 (.5597583)
Construction materials
-1.748438*** (.6451326)
-1.8427 *** (.661591)
Light industry
-.9461021*** (.3357283)
-1.122641*** (.3374742)
Food industry .8116791*** (.3107219)
.8844425 *** (.3141468)
Workers ownership
(.5302774) .3298891
Managers .6611196 (.4943387)
State .0459265 (.4322602)
Physical entities
.4110196 (.3146988)
Ukrainian companies
-.371814 (.3498654)
Foreign companies
.5729147 ** (.2739975)
R2 0.0671 0.1121 0.4654 0.5010 In parentheses are standard errors; *, **, *** mean 10%, 5% and 1% significance level respectively.
32
It could be concluded from Table 5 that FDI influence is positive and
significant for all model variations. Moreover, regional dummies are not
significant that suggests no significant difference among Kyiv, Kharkiv,
Odesa and Lviv regions. As for the industry dummies, labor productivity is
lower in metal processing (S2), construction materials (S4) and light industry
(S5) and higher in food industry (S6). Among ownership dummies, only the
foreign ownership dummy is significant and has a positive impact. Foreign
owned firms have higher labor productivity. So, we could suggest that our
zero hypothesis is rejected statistically.
In order to test our second hypothesis we estimated the model from
Equation 7. The results are in Table 6. The FDI dummy is significant and
positive, which suggests that H0 is econometrically incorrect. Expansion in
the export volume depends on labor. Regional variables are not significant,
which could signify the absence of regional differences. Light industry (S5)
firms have higher export volume. This could indicate that light industry is
more export-oriented than others, because it is labor intensive and Ukraine
has cheap and hill-skilled labor. The coefficients of other industry dummies
are not significant.
Further, we look for the ownership effect. Only two dummies, for the state
(O3) and foreign ownership (O6), are significant. Export orientation of
foreign owners can be explained by the fact that production in Ukraine is
less expensive than in some other countries due to cheap, high-skilled labor
and tax privileges. The significance of state ownership could be a result of
government subsidies. Moreover, state-owned companies could have direct
and indirect advantages. Direct advantages could be explained through tax
holidays, while indirect advantages could imply cheaper prices for gas,
electricity and utilities, which are subsidized by government.
33
Table 6. Regression results for the FDI effect on export.
itExpln itExpln itExpln itExpln
constant 746.0346*** (123.3835)
739.6532*** (128.3026)
952.2325*** (166.672)
844.6346*** (163.6355)
itKln -.065302 (.1685712)
-.0754216 (.177256)
.0375476 (.17749)
.0927884 (.1679689)
itLln 1.053925*** (.2556936)
1.059116*** (.2603795)
.8663591*** (.2800036)
.9558322*** (.2707489)
FDI 46.0487*** (7.622021)
45.65519*** (7.92957)
58.7844*** (10.30754)
52.22984*** (10.11333)
Kyiv region .1402472 (.7403931)
.0381235 (.734901)
-.1492532 (.6920088)
Lviv region .00934 (.7729364)
-.0672211 (.758352)
-.4810016 (.7283313)
Kharkiv region .0473668 (.7619232)
-.1459191 (.7497495)
-.4511203 (.7099927)
Metallurgy industry
.2150633 (.6217436)
.3292866 (.6009516)
Metal processing
.7478811 (.5235675)
.789942 (.4958547)
Wood and paper
.4667229 (1.011058)
.315329 (.9461807)
Construction materials
.1945991 (1.184627)
.725286 (1.143566)
Light industry 1.660828 *** (.6113901)
1.410261** (.5931065)
Food industry -.9703221 * (.5672528)
-.8042281 (.543708)
Scale -.0026378 (.0595256)
-.0159139 (.0576004)
Workers ownership
.0152263 (.5693204)
Managers .0744125 (.8567416)
State 1.135162 (.7350317)
Physical entities
.5281672 (.5320371)
Ukrainian companies
.9791046 ( .5998799)
Foreign companies
2.082235*** (.4919346)
R2 0.3202 0.3215 0.4126 0.5036
In parentheses are standard errors; *, **, *** mean 10%, 5% and 1% significance level respectively.
Finally, we test for spillovers influence on non-FDI firm’s performance.
Model 3 is described by equation 8. We estimate FDI intensity (spillover)
effect on non-FDI firms performance, measured as labor productivity.
According to the results in Table 7, the spillover variable (FDI intensity) is
positive and significant at the 1% level. We could suggest that positive FDI
spillovers exist, but their effect is comparatively low because of small
volume of FDI in Ukraine. Furthermore, firms owned by other Ukrainian
companies (O5) operate worse than firms with other ownership types. This
34
could be explained by competition. Business rivals buy shares each other to
have better access to raw materials Non-FDI firms have lower labor
productivity in metal processing (S2) and wood industries (S3). At the same
time, the metallurgy industry (S1) creates positive externalities.
Table 7. Regression results for the spillovers effect on labor productivity.
it
it
LY
ln it
it
LY
ln it
it
LY
ln it
it
LY
ln
constant .8387315 *** (.2503692)
.9339324 *** (.2723761)
.8037184 *** (.2917585)
.7770575 *** (.2837662)
it
it
L
Kln
.1707783 ** (.0788923)
.17281 ** (.0795979)
.2292445 *** (.0755159)
.2267631 *** (.0753785)
spillover .0029564 *** (.0007592)
.0029796 *** (.0007643)
.0022251 *** (.0007798)
.0023776*** (0007746)
Workers ownership
-.2742787 (.2454183)
-.155701 (.2276825)
Managers .0620174 (.4177473)
.1221477 (.3963747)
State -.2565311 (.5136659)
.4522285 (.4829886)
Physical entities .0554883 (.2879311)
.0297117 (.2631004)
Ukrainian companies
-.5306005* (.3026857)
-.5987747** (.2797578)
Foreign companies
.7990214 (.9661738)
.0781827 (.8985468)
Metallurgy industry
.5885331 * (.3363792)
.5185387 (.334257)
Metal processing
-.7742763*** (.2774144)
-.8045659*** (.2710271)
Wood and paper
-.8716325** (.4409653)
-.8462005** (.4235343)
Construction materials
-.0539755 (.3273139)
-.1288253 (.3251478)
Light industry -.4108193 (.3406818)
-.4664606 (.3348827)
Food industry .7299265 (.2758278)
.6428868 (.2670187)
R2 0.0898 0.1091 0.2734 0.2504
In parentheses are standard errors; *, **, *** mean 10%, 5% and 1% significance level respectively.
Finally, our model 4 where performance is estimated as export volumes is in
the Equation 9.
The results of the model are depicted in Table 8. The spillover variable is
statistically significant which implies the rejection of Ho for the model 4.
The coefficient by spillover variable is low and smaller less than the
coefficient of the labor variable, which is also significant. While industry
35
dummies are not significant, firms owned by state (O3) and other physical
entities (O4) do benefit from the type of ownership.
Table 8. Regression results for spillovers influence on volumes of export.
itExpln itExpln itExpln itExpln
constant -1.986797*
(1.081524) -2.522509 ** (1.050605)
-2.167826** (1.012107)
-2.589027 (1.160966)
itLln 1.174555*** (.1607732)
1.169415*** (.1585468)
1.164557 *** (.1541413)
1.22317 *** (.1694424)
spillover .0029216 * (.0016981)
.0028104** (.00142)
.0028077 * (.0014433)
.0032366* (.0017117)
Workers ownership
.3753603 (.610116)
Managers .7417539 (.9254058)
State 1.190222 (.7258599)
1.324532* (.7662441)
Physical entities
1.051411* (.5439513)
1.233253** (.5864281)
Ukrainian companies
.8572546 (.6658557)
1.129888 (.7260139)
Metallurgy industry
.7174181 (.7039667)
.9029099 (.5839055)
.9589779 (.5877421)
.4705539 (.7169181)
Metal processing
-.3159037 (.5702525)
-.6071945 (.5736704)
Wood and paper
.9759838 (2.033107)
1.248586 (2.006702)
Construction materials
-1.185085 (1.168831)
-1.144203 (1.191918)
Light industry .7235154 (.7583744)
1.064151* (.6428224)
.9691074 (.6497758)
.5397828 (.7755485)
Food industry -.283287 (.7596769)
-.550465 (.7884481)
R2 0.3659 0.3932 0.3535 0.4082
In parentheses are standard errors; *, **, *** mean 10%, 5% and 1% significance level respectively.
36
S e c t i o n 4
4. CONCLUSIONS AND SUGGESTION FOR FUTURE RESEARCH
The issue of FDI and spillovers is highly appealing for research, because
FDI could have not only positive, but also negative externalities. FDI also
has direct and indirect impacts. Direct FDI effects measure differences in
firm indicators between firms with and without FDI. Indirect effects are
spread through interactions between foreign and domestic firms. There are
five main indirect effects found in relevant literature: technology transfer,
catch-up, competition effect, foreign linkage effect and training effect.
Using unpublished micro-level annual data for 292 firms during 1998-99, we
test for statistical significance of FDI impact on labor productivity (Model
1) and export volume (Model 2). Furthermore, we investigate spillover
effect in Models 3-4.
The results reported in the paper imply that the presence of FDI has a
positive influence on both labor productivity and exports. Regional variables
are not significant that could imply the absence of differences for Kyiv,
Kharkiv, Odesa and Lviv regions. There is also a small spillover effect that
signifies positive impact of FDI environment on such non-FDI firms’ labor
productivity and export volumes. Hence, small volumes of FDI cannot have
much influence on performance.
As for industry dummies in Model 1, firms from metal processing,
construction materials and light industry have lower, while enterprises in
food industry have higher labor productivity. We can suggest from model 2
that light industry companies export more then firms from other industries.
According to Model 3, metal processing and wood industries have lower
labor productivity for non-FDI firms than others industries. At the same
time, metallurgy industry creates positive externalities.
37
Among ownership dummies in Model 1 and 2, foreign ownership dummies
are significant and have a positive impact. Export orientation of foreign
owners can be explained by a greater efficiency of foreign owners. The
significance of state ownership in Model 2 could be a result of government
subsidies, tax privileges or other policies. According to Model 3 results,
firms owned by other Ukrainian companies operate worse than firms with
other types of ownership.
Despite popularity in other transition countries, the topic of the effect of
FDI on firm performance is rarely described in Ukrainian applications. As
for the future research, ideally, it would be challenging to assess the effects
of FDI volumes on firm performance. Moreover, the sample of 292 firms
may not accurately describe the real situation and it is important to expand
the sample. Also, it would be advantageous to estimate the effects of
industry and regional spillovers separately. Finally, output per worker and
export could not be the best indicators of firm’s performance. We expect
value added and value added per worker to be the better measures.
WORKS CITED
Blomstrom Magnus. 1999. “Internationalization and Growth: Evidence from Sweden”, Stockholm School of Economic Working Paper No. 375.
----. 1989 “Foreign investment and Spillovers”, Routledge.
----, Ari Kokko. 1997. “How Foreign Investment Affects Host Countries?”, The World Bank Policy Research Working Paper, No.1745.
-----, -----, and Steven Globerman. 1999. “The Determinants of Host Country Spillovers from Foreign Direct Investment: Review and Synthesis of the Literature”. Stockholm School of Economic Working Paper No. 339.
Davies, Howard. 1977. “Technology Transfer Through Commercial Transactions”, Journal of Industrial Economics, Vol. 26, No. 2, pp.161-175.
Djankov, Simeon, Hoekman Bernard. 1998. “Avenues of Technology Transfer: Foreign Investment and Productivity Change in Czech Republic”, Center for Economic Policy Research, Discussion Paper, No.1883.
Dyker, David A.. 1999. Foreign Direct Investment and Technology Transfer in the Former Soviet Union”, Cheltenham, U.K. and Northampton, Mass.: Elgar; Distributed by American International Distribution Corporation, Williston, Vt.
Estrin S. Rosevear A. 1999. “Enterprise performance and ownership: The case of Ukraine”, European Economic Review, 43 pp.1125-1136.
Findlay, Ronald. 1978.“Some Aspects of Technology Transfer and Direct Foreign Investment”, The American Economic Review, vol.68, No.2, pp.275-279.
Glass Amy Jocelyn and Kamal Saggi. (1998) Intellectual property rights and Foreign Direct Investments. Mimeo. Ohio State University.
Green, William H. 2000. Econometric Analysis. Prentice Hall.
Grossman, G., and Helpman, E. 1991a. Innovation and Growth in the Global Economy. MIT Press.
Gujarati, Damodar N. 1995. Basic Econometrics, Mc Graw Hill, Inc.
Johnson, Jack and John DiNardo. 1997 Econometric Methods. McGraw-Hill.
Herasymenko, V. 2000. “Foreign Direct Investment and its Role for Transition Countries”, materials of international conference “Crossboarder Capital Flows in Transition Economies”, mimeo., EERC-Kyiv.
Institute of reform (2000). Å ê î í î ì ³ ÷ í ³ åñå: ðåéòèíã ³íâåñòèö³éíî¿ ïðèâàáëèâîñò³ ðåã³îí³â Óêðà¿íè ó 1- ìó ï³âð³÷÷³ 2000 ðîêó. Vol 4.
Jensen, Richard, Marie Thursby.1987. “A Decision
2
Theoretic Model of Innovation, Technology Transfer, and Trade”, Review of Economic Studies, vol.54, No.4, pp.631-647.
Kinoshita, Yuko. 1998. Technology Transfer through Foreign Direct Investments, mimeo., CERGE
_____. 2000. R&D and Technology Spillovers via FDI: Innovation and Absorptive Capacity, mimeo., CERGE
Konings, Josef. 2000. The Effects of Foreign Direct Investments, on Domestic Firms: Evidence from Firm Level Panel Data in Emerging Economies. Center for Economic Policy Research, Discussion Paper, No.2586.
Kudina, Alina. 1999. “The Motives for Foreign Direct Investment in Ukraine, Master’s Thesis, mimeo., EERC-Kyiv.
Loo, Frances Van.1977. “The Effect of Direct Foreign Investment on Investment in Canada”, Review of Economics and Statistics, vol.59, No.4, pp.474-481.
Markusen, James, Rutherford F. Thomas, David Tarr. 2000. “Foreign Direct Investment in Services and Market for Expertise?”, NBER Working Paper, No.W7700
Mykytiv, A. 2000. “Does Foreign Direct Investment Have Positive Impact on Economic Growth in Transition Economies? A Cross-Country Study”, materials of international conference “Crossborder Capital
Flows in Transition Economies”, mimeo., EERC-Kyiv.
Oleksiv, M. 2000. “What Factors Affect Foreign Direct Investment Flow in Ukraine”, materials of international conference “Crossborder Capital Flows in Transition Economies”, mimeo., EERC-Kyiv.
Ponomareva, N. 2000. “Foreign Direct Investment in Russia: Effects on Productivity”, materials of international conference“ Crossborder Capital Flows in Transition Economies”, mimeo., EERC-Kyiv.
Ramachandran, Vijaya.1993. “Technology Transfer, Firm Ownership, and Investment in Human Capital”, Review of Economics and Statistics, vol.75, No.4, pp.664-670.
Sinn, Hans Werner and Weichenrieder Alfons 1997. Foreign Direct Investment, Political Resentment and the Privatization Process in Eastern Europe, Economic Policy: A European Forum; 0(24), April 1997, pp177-98.
Yegorov Igor, 1999. Foreign direct investments in Ukraine: First results, Tendencies and Prospects”, Foreign Direct Investments and Technology Transfer in the Former Soviet Union. pp.155-188
The Law of Ukraine “On Foreign Investments”. 1992,
The Law “On The Foreign Investing Regime”, 1996,
The Law “On The State Program of Encouraging of Foreign Investments in Ukraine”,1993,
3
APPENDICES
A1. Graphs
Regional Distribution of Dataset
Kyiv region30%
Lviv region31%
Odesa region9%Kharkiv region
30%
Industry distribution in dataset
Metallurgy8%
Light10%
Food26%
Others22%
Construction materials
9%
Wood and Paper5%
Metal processing20%
4
A2. Stata 6.0 do-file program
use "e:\stata\panel.dta", clear log using e:\stata\results.log, replace iis eerc tis year /*Variables*/ gen lnk=log(bfa) /*Capital:*/ gen lnl=log(lab) /*Labor*/ gen spil=rinv*ispil/* FDI intensity : %industry*%region*/ gen lny_l=log(y/lab) /*Production per worker*/ gen lns_l=log(a/lab) /*Sales per worker*/ gen lnk_l=log(bfa/lab) /*Capital per worker*/ gen lbar=bar/a /*% of Barter*/ gen lexp=log(exp) /* Export*/ gen lexp_l=log(exp/lab) /* Export per worker*/ gen exp_s=exp/a /* % of export in sales*/ egen scale1 = mean(y), by (i1-i6) gen scale = y/scale1 /*Economy of Scale:*/ /*Hypothesis 1*/ xtprobit fdi lexp predict fdi_n, xb xtreg lny_l lnk_l fdi_n r1-r3 i1-i6 o1-o6, re xthaus xtreg lny_l lnk_l fdi_n, re xtreg lny_l lnk_l fdi_n r1-r3, re xtreg lny_l lnk_l fdi_n r1-r3 i1-i6, re /*Hypot hesis 2*/ xtprobit fdi lny_l predict fdi1_n, xb xtreg exp_s lnk lnl fdi1_n r1-r3 i1-i6 scale o1-o6, re xthaus xtreg exp_s lnk lnl fdi1_n, re xtreg exp_s lnk lnl fdi1_n r1-r3, re xtreg exp_s lnk lnl fdi1_n r1-r3 i1-i6, re xtreg exp_s lnk lnl fdi1_n r1-r3 i1-i6 scale, re /*Hypothesis 3*/ xtreg lny_l lnk_l i1-i6 o1-o6 spil if fdi==0, re xthaus xtreg lny_l lnk_l spil if fdi==0, re xtreg lny_l lnk_l o1-o6 spil if fdi==0, re xtreg lny_l lnk_l i1-i6 spil if fdi==0, re /*Hypothesis 4*/ xtreg lexp lnl i1-i6 spil if fdi==0, re xtreg lexp lnl i1 i5 o3-o5 spil if fdi==0, re xtreg lexp lnl i1 i5 spil if fdi==0, re xtreg lexp lnl i1-i6 o1-o5 spil if fdi==0, re xthaus log close
5
A3. Hausman specification tests.
A3.1. Hausman specification test for Model 1.
R-sq: within = 0.1875 Obs per group: min = 1 between = 0.4924 avg = 1.7 overall = 0.5010 max = 2 Random effects u_i ~ Gaussian Wald chi2(17) = 122.46 corr(u_i, X) = 0 (assumed) Prob > chi2 = 0.0000 ------------------------------------------------------------------------------ lny_l | Coef. Std. Err. z P>|z| [95% Conf. Interval] ---------+-------------------------------------------------------------------- lnk_l | -.0483727 .0878096 -0.551 0.582 -.2204764 .1237309 fdi_n | .7737273 .1398052 5.534 0.000 .4997141 1.047741 r1 | .0755219 .4087269 0.185 0.853 -.7255681 .8766119 r2 | -.3236778 .4304532 -0.752 0.452 -1.167351 .5199949 r3 | .1697789 .420667 0.404 0.687 -.6547133 .9942711 i1 | .1002837 .3539814 0.283 0.777 -.5935072 .7940745 i2 | -1.105147 .2791934 -3.958 0.000 -1.652356 -.5579379 i3 | .069621 .5597583 0.124 0.901 -1.027485 1.166727 i4 | -1.8427 .661591 -2.785 0.005 -3.139394 -.5460053 i5 | -1.122641 .3374742 -3.327 0.001 -1.784078 -.4612036 i6 | .8844425 .3141468 2.815 0.005 .268726 1.500159 o1 | .5302774 .3298891 1.607 0.108 -.1162934 1.176848 o2 | .6611196 .4943387 1.337 0.181 -.3077665 1.630006 o3 | .0459265 .4322602 0.106 0.915 -.8012879 .8931409 o4 | .4110196 .3146988 1.306 0.192 -.2057786 1.027818 o5 | -.371814 .3498654 -1.063 0.288 -1.057538 .3139096 o6 | .5729147 .2739975 2.091 0.037 .0358893 1.10994 _cons | 3.361068 .6017184 5.586 0.000 2.181722 4.540415 ---------+-------------------------------------------------------------------- sigma_u | .98669983 sigma_e | .29996958 rho | .91539557 (fraction of variance due to u_i) ------------------------------------------------------------------------------ Hausman specification test ---- Coefficients ---- | Fixed Random lny_l | Effects Effects Difference ---------+----------------------------------------- lnk_l | -.3014501 -.0483727 -.2530774 fdi_n | .7501185 .7737273 -.0236089 Test: Ho: difference in coefficients not systematic chi2( 2) = (b-B)'[S^(-1)](b-B), S = (S_fe - S_re) = 2.92 Prob>chi2 = 0.2319
6
A3.2. Hausman specification test for Model 2. R-sq: within = 0.0063 Obs per group: min = 1 between = 0.2132 avg = 1.8 overall = 0.2009 max = 2 Random effects u_i ~ Gaussian Wald chi2(19) = 71.78 corr(u_i, X) = 0 (assumed) Prob > chi2 = 0.0000 ------------------------------------------------------------------------------ exp_s | Coef. Std. Err. z P>|z| [95% Conf. Interval] ---------+-------------------------------------------------------------------- lnk | -.034229 .038994 -0.878 0.380 -.110656 .0421979 lnl | .1006678 .0599572 1.679 0.093 -.0168461 .2181818 fdi1_n | 3.592282 1.787189 2.010 0.044 .0894569 7.095108 r1 | -.0100571 .1532398 -0.066 0.948 -.3104016 .2902873 r2 | .1496963 .1549547 0.966 0.334 -.1540094 .453402 r3 | -.0783437 .146822 -0.534 0.594 -.3661096 .2094222 i1 | -.0429657 .1591218 -0.270 0.787 -.3548387 .2689074 i2 | .0285692 .1287305 0.222 0.824 -.2237378 .2808762 i3 | -.0286531 .1861884 -0.154 0.878 -.3935757 .3362695 i4 | -.0427706 .1515755 -0.282 0.778 -.3398531 .2543118 i5 | .6408093 .1502053 4.266 0.000 .3464123 .9352062 i6 | -.1661267 .1166942 -1.424 0.155 -.3948432 .0625898 scale | -.0368867 .0193662 -1.905 0.057 -.0748439 .0010704 o1 | -.1890413 .1168057 -1.618 0.106 -.4179762 .0398937 o2 | -.0783898 .2076067 -0.378 0.706 -.4852914 .3285118 o3 | .7473159 .2625794 2.846 0.004 .2326697 1.261962 o4 | .1578111 .1339226 1.178 0.239 -.1046724 .4202947 o5 | .0147235 .1367398 0.108 0.914 -.2532815 .2827286 o6 | .2566897 .1210695 2.120 0.034 .0193978 .4939816 _cons | 58.1132 28.99299 2.004 0.045 1.287987 114.9384 ---------+-------------------------------------------------------------------- sigma_u | .5100323 sigma_e | .42710261 rho | .58780519 (fraction of variance due to u_i) ------------------------------------------------------------------------------ Hausman specification test ---- Coefficients ---- | Fixed Random exp_s | Effects Effects Difference ---------+----------------------------------------- lnk | -.1477823 -.034229 -.1135532 lnl | .0220697 .1006678 -.0785982 fdi1_n | 4.420467 3.592282 .8281843 scale | -.0756615 -.0368867 -.0387748 Test: Ho: difference in coefficients not systematic chi2( 4) = (b-B)'[S^(-1)](b-B), S = (S_fe - S_re) = 1.99 Prob>chi2 = 0.7368
7
A3.3. Hausman specification test for Model 3. R-sq: within = 0.0001 Obs per group: min = 1 between = 0.3081 avg = 1.9 overall = 0.2734 max = 2 Random effects u_i ~ Gaussian Wald chi2(14) = 83.51 corr(u_i, X) = 0 (assumed) Prob > chi2 = 0.0000 ------------------------------------------------------------------------------ lny_l | Coef. Std. Err. z P>|z| [95% Conf. Interval] ---------+-------------------------------------------------------------------- lnk_l | .2292445 .0755159 3.036 0.002 .0812361 .3772529 i1 | .5885331 .3363792 1.750 0.080 -.0707581 1.247824 i2 | -.7742763 .2774144 -2.791 0.005 -1.317999 -.230554 i3 | -.8716325 .4409653 -1.977 0.048 -1.735909 -.0073563 i4 | -.0539755 .3273139 -0.165 0.869 -.6954989 .587548 i5 | -.4108193 .3406818 -1.206 0.228 -1.078543 .2569047 i6 | .7299265 .2758278 2.646 0.008 .189314 1.270539 o1 | -.155701 .2276825 -0.684 0.494 -.6019505 .2905484 o2 | .1221477 .3963747 0.308 0.758 -.6547325 .8990278 o3 | .4522285 .4829886 0.936 0.349 -.4944118 1.398869 o4 | .0297117 .2631004 0.113 0.910 -.4859557 .545379 o5 | -.5987747 .2797578 -2.140 0.032 -1.14709 -.0504595 o6 | .0781827 .8985468 0.087 0.931 -1.682937 1.839302 spil | .0022251 .0007798 2.854 0.004 .0006968 .0037534 _cons | .8037184 .2917585 2.755 0.006 .2318822 1.375555 ---------+-------------------------------------------------------------------- sigma_u | 1.1252245 sigma_e | .50983428 rho | .82967198 (fraction of variance due to u_i) ------------------------------------------------------------------------------ . xthaus Hausman specification test ---- Coefficients ---- | Fixed Random lny_l | Effects Effects Difference ---------+----------------------------------------- lnk_l | .0200656 .2292445 -.2091789 Test: Ho: difference in coefficients not systematic chi2( 1) = (b-B)'[S^(-1)](b-B), S = (S_fe - S_re) = 1.78 Prob>chi2 = 0.1815
8
A3.4. Hausman specification test for Model 4.
R-sq: within = 0.0550 Obs per group: min = 1 between = 0.4575 avg = 1.7 overall = 0.4082 max = 2 Random effects u_i ~ Gaussian Wald chi2(13) = 69.01 corr(u_i, X) = 0 (assumed) Prob > chi2 = 0.0000 ------------------------------------------------------------------------------ lexp | Coef. Std. Err. z P>|z| [95% Conf. Interval] ---------+-------------------------------------------------------------------- lnl | 1.22317 .1694424 7.219 0.000 .8910686 1.555271 i1 | .4705539 .7169181 0.656 0.512 -.9345799 1.875688 i2 | -.6071945 .5736704 -1.058 0.290 -1.731568 .5171789 i3 | 1.248586 2.006702 0.622 0.534 -2.684478 5.18165 i4 | -1.144203 1.191918 -0.960 0.337 -3.480319 1.191913 i5 | .5397828 .7755485 0.696 0.486 -.9802645 2.05983 i6 | -.550465 .7884481 -0.698 0.485 -2.095795 .994865 o1 | .3753603 .610116 0.615 0.538 -.820445 1.571166 o2 | .7417539 .9254058 0.802 0.423 -1.072008 2.555516 o3 | 1.324532 .7662441 1.729 0.084 -.1772786 2.826343 o4 | 1.233253 .5864281 2.103 0.035 .0838747 2.382631 o5 | 1.129888 .7260139 1.556 0.120 -.2930733 2.552849 spil | .0032366 .0017117 1.891 0.059 -.0001182 .0065914 _cons | -2.589027 1.160966 -2.230 0.026 -4.864479 -.3135745 ---------+-------------------------------------------------------------------- sigma_u | 1.5751417 sigma_e | 1.1523955 rho | .65135595 (fraction of variance due to u_i) ------------------------------------------------------------------------------ . xthaus Hausman specification test ---- Coefficients ---- | Fixed Random lexp | Effects Effects Difference ---------+----------------------------------------- lnl | 2.24936 1.22317 1.026191 Test: Ho: difference in coefficients not systematic chi2( 1) = (b-B)'[S^(-1)](b-B), S = (S_fe - S_re) = 0.78
Prob>chi2 = 0.3774
9
Ukrainian Industrial Enterprise Survey ‘2000 A4. Questionnaire. Total information about enterprise
À. Ownership 1. Specify the ownership of your enterprises
1. State-owned enterprise 2. Non-state owned enterprise, but it was state-owned before (till (year)) 3. Non-state owned enterprise, it has never been state-owned
2. Specify the legal form of your enterprise 1. Closed joint stock company 2. Open joint stock company 3. Cooperative 4. Partnership 5. Collective enterprises 6. Leased enterprise 7. Individual ownership 8. Joint venture 9. Other (please, specify)
3. If your enterprise is a joint stock company of any type, how are the shares distributed among the shareholders?
workers % managers % government % other physical entities % other Ukrainian companies % other foreign companies % other %
B. Size of enterprise 1. What was the number of workers on floor in _______? ______________________ 2. What was the number of workers on forced leave in _______? _________________ C. Industry What portion of your output belongs to the following sectors of industry? 1 ______________________ metallurgy, energy, chemical industry, coal industry 2 ______________________ machine building 3 ______________________ wood processing 4 ______________________ construction materials 5 ______________________ light industry 6 ______________________ food processing 7 ______________________ printing 8 ______________________ other Please, mention four main types of output produced by your enterprise: à._____________________________________________________ á._____________________________________________________ â._____________________________________________________ ã._____________________________________________________ D. Average Per Sent of Capacity Utilization in 1999 ________%
E. How did FDI change in 2000 compared to 1999?
1. increse 0. The same -1. Decrease 4. Never received FDI 5. DK